Downloads · 30 days
884
46% of all-time downloads
sakamakismile/DeepSeek-V4-Flash-DSpark-support-ds4-GGUF
DeepSeek-V4-Flash-DSpark-support-ds4-GGUF is a text generation model from sakamakismile. Use it when you need the model to write or continue text. It is set up for gguf. The card lists the license as mit.
A standalone DSpark speculative-decoding draft ("support") model, packaged as a single 5.6 GiB GGUF for the ds4 engine. It is the drafter only — it does not generate on its own. Pair it with a DeepSeek-V4-Flash main-m…
Downloads · 30 days
884
46% of all-time downloads
All-time downloads
1.9K
Public
Repo size
6 GB
Likes
5
Public
Click a slice to open those files.
.gguf6 GB · 100%
From the Hugging Face model README
A standalone DSpark speculative-decoding draft ("support") model, packaged as a single 5.6 GiB GGUF for the ds4 engine. It is the drafter only — it does not generate on its own. Pair it with a DeepSeek-V4-Flash main-model GGUF and ds4 proposes-and-verifies, committing only accepted prefixes.
This file is what ./download_model.sh dspark-support produces, but pre-built: you can
skip the ~167 GB DSpark source download and the local conversion.
DSpark is DeepSeek's official auxiliary draft module for DeepSeek-V4-Flash: three MTP (multi-token-prediction) stages that read the main model's hidden states at layers 40–42 and propose a block of future tokens, with a Markov head for the block continuation. ds4 verifies every proposal against the main model, so accepted tokens are exactly what the main model would have produced — speculative decoding is a speed optimization, not a quality change.
| File | Size | SHA-256 |
|---|---|---|
DeepSeek-V4-Flash-DSpark-support.gguf | 5,989,114,272 B (5.6 GiB) | 8b3adf5942bec22ae2ea867cd7079cf13530ba83ffcffaf00f5de48664a1a34e |
general.architecture = deepseek4-dspark.block_size = 5, markov_rank = 256, n_layers = 3,
target_layer_ids = [40, 41, 42], noise_token_id = 128799.deepseek4-quantize DSpark builder): routed experts
ffn_down Q2_K / ffn_gate,ffn_up IQ2_XXS; attention, shared experts and main_proj
Q8_0; norms F32; hyper-connection tensors F16.Requires a DeepSeek-V4-Flash main-model GGUF (e.g. ds4flash.gguf) and a recent ds4 build.
./ds4 -m ds4flash.gguf \
--mtp DeepSeek-V4-Flash-DSpark-support.gguf \
--dspark --temp 0
Notes:
--dspark activates the runtime; --dspark-confidence 0.9 is the default acceptance
threshold (--dspark-confidence 0 forces fixed five-token blocks for diagnostics).--temp 0); sampled and reasoning/think modes
disable it.deepseek-ai/DeepSeek-V4-Flash-DSpark
(MIT) — the DeepSeek-V4-Flash checkpoint plus the DSpark drafter module. Only the
drafter tensors (the mtp.* stages + Markov head, checkpoint shards 46–48) were used.gguf-tools/deepseek4-quantize DSpark support builder, from
antirez/ds4 commit fc9efd1 ("Add DSpark speculative
decoding"). No engine code was modified to produce this file.block_size, markov_rank, target_layer_ids, noise_token_id).This is a straightforward format conversion of MIT-licensed weights. It has been built and
metadata-verified; end-to-end acceptance/throughput on your specific hardware should be
confirmed with ds4's make dspark-acceptance fixture.
MIT, inherited from the base model. This is a derivative of DeepSeek's checkpoint and adds no additional restrictions. Not affiliated with or endorsed by DeepSeek or the ds4 author; all credit for the DSpark method and weights goes to DeepSeek, and to antirez for the ds4 engine and its DSpark GGUF builder.